HomePricingBlogsDocs

Gemma 4 models

42 Gemma 4 models on Hugging Face, from 420M to 32.7B parameters. At the precision each one is published in, the smallest needs about 0.9 GB of VRAM (gemma-4-26B-A4B-it-assistant, cheapest live fit: RTX 5060 Ti) and the largest about 73.1 GB (gemma-4-31B, cheapest live fit: A100). The cheapest way to run gemma-4-26B-A4B-it-assistant is $0.110/hr.

Gemma 4 models

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
gemma-4-26B-A4B-it-assistant420MBF160.9 GBRTX 5060 Ti$0.110/hr
gemma-4-12B-it-assistant423MBF160.9 GBRTX 5060 Ti$0.110/hr
gemma-4-12B-it-qat-q4_0-unquantized-assistant423MBF160.9 GBRTX 5060 Ti$0.110/hr
gemma-4-26B-A4B-it-DFlash430MBF161.0 GBRTX 5060 Ti$0.110/hr
gemma-4-31B-it-assistant470MBF161.0 GBRTX 5060 Ti$0.110/hr
gemma-4-31B-it-qat-q4_0-unquantized-assistant470MBF161.0 GBRTX 5060 Ti$0.110/hr
gemma-4-31B-it-DFlash1.5BBF163.4 GBRTX 5060 Ti$0.110/hr
gemma-4-E4B-it-W4A16-AutoRound-GPTQ4.0BBF168.9 GBRTX 5060 Ti$0.110/hr
gemma-4-E2B-it-unsloth-bnb-4bit5.1BBF1611.5 GBRTX 5060 Ti$0.110/hr
gemma-4-E2B-it5.1BBF1611.5 GBRTX 5060 Ti$0.110/hr
gemma-4-E2B-it-qat-w4a16-ct5.6BBF1612.4 GBRTX 5060 Ti$0.110/hr
Gemma-4-E4B-DECKARD-HERETIC-NVFP46.2BBF1613.9 GBRTX 5060 Ti$0.110/hr
gemma4-e4b-claims-comparison7.9BBF1617.7 GBRTX A5000$0.176/hr
gemma-4-E4B-it-OBLITERATED8.0BF1617.9 GBRTX A5000$0.176/hr
gemma-4-E4B-it-unsloth-bnb-4bit8.0BBF1617.9 GBRTX A5000$0.176/hr
gemma-4-E4B8.0BBF1617.9 GBRTX A5000$0.176/hr
gemma-4-E4B-it8.0BBF1617.9 GBRTX A5000$0.176/hr
gemma-4-E4B-it-uncensored-heretic8.0BBF1617.9 GBRTX A5000$0.176/hr
gemma-4-E4B-it-AWQ-INT48.2BF1618.3 GBRTX A5000$0.176/hr
gemma-4-E4B-it-qat-w4a16-ct8.7BBF1619.5 GBRTX A5000$0.176/hr
Jev-Omni12.0BBF1626.7 GBRTX 4080 Super$0.338/hr
gemma-4-12B12.0BBF1626.7 GBRTX 4080 Super$0.338/hr
gemma-4-12B-it12.0BBF1626.7 GBRTX 4080 Super$0.338/hr
gemma-4-12B-it-qat-q4_0-unquantized12.0BBF1626.7 GBRTX 4080 Super$0.338/hr
humanizer12.0BBF1626.7 GBRTX 4080 Super$0.338/hr
Gemma-4-12B-OBLITERATED12.0BBF1626.7 GBRTX 4080 Super$0.338/hr
gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau212.0BF1626.7 GBV100$0.187/hr
gemma-4-12B-it-FP8-Dynamic13.0BF8_E4M314.5 GBRTX 5060 Ti$0.110/hr
Gemma-4-31B-IT-NVFP420.9BBF1646.6 GBRTX A6000$0.363/hr
GEV-26B-Decide25.8BBF1657.7 GBA100$1.21/hr
gemma-4-26B-A4B-it25.8BBF1657.7 GBA100$1.21/hr
rune-26b-a4b-GGUF25.8BBF1657.7 GBA100$1.21/hr
Orion-26B-A4B-v125.8BBF1657.7 GBA100$1.21/hr
gemma-4-26B-A4B-it-qat-q4_0-unquantized26.5BBF1659.3 GBA100$1.21/hr
gemma-4-26B-A4B-it-FP8-dynamic26.5BF8_E4M329.7 GBRTX 4080 Super$0.338/hr
gemma-4-31B-it31.3BBF1669.9 GBA100$1.21/hr
gemma-4-31B-it-FP8-block31.3BF8_E4M335.0 GBRTX 4090$0.485/hr
gemma-4-31B-it-FP8-dynamic31.3BF8_E4M335.0 GBRTX 4090$0.485/hr
Artemis-31B-v1.231.3BBF1669.9 GBA100$1.21/hr
OTel-2.0-LLM-31B-IT32.1BBF1671.8 GBA100$1.21/hr
gemma-4-31B32.7BBF1673.1 GBA100$1.21/hr
gemma-4-31B-it-uncensored32.7BBF1673.1 GBA100$1.21/hr

VRAM is for the precision the model is published in, with the same overhead and an 8,192-token context assumed on every page; see the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.

More from Google

All modelsBrowse the GPU Index

Submit the job. Everything after that is ours.

Sign up in 60 seconds. Pay for the GPU minutes you actually use.

Aquanode LogoAquanode

Run AI without spending any time on infrastructure.

© 2026 Aquanode. All rights reserved.

All trademarks, logos and brand names are the property of their respective owners.